Directories and arguments

[M K]

Hello, You mention that the data when running the submission will already be prepared with these scripts from: https://github.com/physionetchallenges/python-example-2025?tab=readme-ov-file#how-do-i-create-data-for-these-scripts Is that correct? Also our submission failed in train_model.py and the error showed the train_model.py --data-path argument. Instead of "training_data", you provided "data" folder. Everything works fine in our local machines when following docker and scripts that you provide here: https://github.com/physionetchallenges/python-example-2025?tab=readme-ov-file#how-do-i-run-these-scripts-in-docker ). Does this "data" folder instead of "training_data" contain prepared data? Maybe you could list all of the exact commands with arguments when running the submission or is it the same as in python-example-2025 except for the --data-folder argument? Or should we explicitly write what directory structure we expect in our README? Best, MK

[PhysioNet Challenge]

Dear MK,

Yes, that’s right: the data will already be there, and you don’t need to prepare it yourself.

> Also our submission failed in train_model.py and the error showed the train_model.py --data-path argument. Instead of "training_data", you provided "data" folder. Everything works fine in our local machines when following docker and scripts that you provide here: https://github.com/physionetchallenges/python-example-2025?tab=readme-ov-file#how-do-i-run-these-scripts-in-docker ).

Hm, I can see how the instructions could be a little ambiguous here.

Paths like /training_data and /holdout_data are meant to stand for arbitrary locations in the container where you can mount the data. It’s better not to hard-code these at all.

In your code, please use the arguments supplied to the train_model, load_model, and run_model functions. So, no matter what “model_folder” is, save the model there. No matter where “data_folder” is, read the data there. No matter what “record” may be, use it and return an output. That way, we can make the code as reusable as possible.

Best,

James

[Guilherme Santos]

Hello PhysioNet Team,

Is there still time to submit one or two more versions of the code and hopefully see better scores on the leaderboard?

--
You received this message because you are subscribed to the Google Groups "physionet-challenges" group.
To unsubscribe from this group and stop receiving emails from it, send an email to physionet-challe...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/physionet-challenges/002fcf76-7018-4275-88f6-1fa7ac3038c0n%40googlegroups.com.

[PhysioNet Challenge]

Oh yes, please do keep submitting!

Best,

James

[Vinay Pai]

We have till midnight eastern to submit?